Thermoformable shin guards — the kind a player softens in hot water and moulds to their own shin — are being asked about far more often than they are being tested. So we ran two candidate samples through a protocol and measured what they did, rather than repeating what the supplier said they would do.
Below is the protocol, the numbers, and the one finding that mattered more than the numbers.
The protocol
Anyone can repeat this. That is the point of writing it down.
- Baseline measurement. Calipers at the thickest point of the shell and, separately, of the EVA layer. Overall length and width at the maximum. Weight on a 0.1 g scale. For a two-piece sample, weigh the parts separately as well as assembled.
- Immersion. Water held at 70 °C. Sample fully submerged. Time to the point where it is hand-mouldable, recorded with a stopwatch.
- Surface temperature on exit. Measured immediately out of the water, on the face that will touch skin. This is a safety measurement, not a process one — see below.
- Moulding. Pressed onto a shin and held until it holds shape. For the two-piece sample, the shell is moulded on its own, with no EVA in between.
- Cooling and re-hardening. Timed at ambient, no forced cooling.
- Re-measurement. Calipers again at the same points. A material that thins under hand pressure is not the same product after moulding as before it.
- Second cycle. Repeat 2–5 on the same sample. "Remouldable" is a claim about the second cycle, not the first.
- Reference sample. A conventional carbon-shell guard measured alongside, as an upper bound on weight and thickness. It is not thermoformable — it is there to give the numbers a scale.
What we measured
All testing at 70 °C, cooled at ambient.
| Parameter | Sample 001 | Sample 002 | Carbon reference |
|---|---|---|---|
| Construction | Two-piece | One-piece | — |
| Shell thickness | 2.0 mm | — | — |
| EVA thickness | 3.0 mm | — | — |
| Total thickness | 4.8 mm | 3.5 mm | — |
| Thickness after moulding | — | 3.5 mm (unchanged) | — |
| Max length | not measured | 130.3 mm | 99.0 mm |
| Max width | not measured | 81.1 mm | 63.7 mm |
| Weight | 33.3 g | 24.0 g | 11.0 g |
| Softening | 70 °C / 30 s | 70 °C / 30 s | — |
| Surface temp out of water | 43 °C | — | — |
| Cooling / re-hardening | 3 min | 3 min | — |
| Remouldable | Yes | Yes | — |
Both samples softened in 30 seconds at 70 °C, moulded by hand, re-hardened on cooling, and did it again on a second cycle. Sample 002 measured the same thickness after moulding as before it.
Finding 1 — 43 °C is the number that matters, not 70 °C
The water is at 70 °C. The sample's surface is 43 °C the moment it comes out.
Thin-walled plastic has very little thermal mass, so it drops into comfortable handling range within seconds of leaving the water. For scale, a hot shower runs at 40–42 °C.
In practice that means a player can take it straight out and press it onto their shin bare-handed — no waiting, no gloves, no protective step in the instructions. For a product moulded at home by a consumer rather than in a workshop, this is the difference between a product and a liability.
If you are evaluating one of these, measure the exit surface temperature, not the water temperature. The water temperature tells you about the process. The surface temperature tells you about the user.
Finding 2 — two constructions, two different weaknesses
This is the part worth your attention, and it is not visible in a photograph.
Sample 001 is two-piece. The shell and the EVA are separate. You soften the shell alone, mould it, and attach the EVA once it has set. The bond can be peeled apart by hand.
We initially took that peelable bond for a defect. Testing showed it is the design intent: with no EVA cushion in between, the shell moulds directly against the shin and the contour is accurate.
Sample 002 is one-piece. Laminated at the factory, immersed and moulded as a single part.
The difference is not only in process — it determines how the product is used:
| Item | Two-piece (001) | One-piece (002) |
|---|---|---|
| Moulding accuracy | Shell against skin, accurate | Through the EVA, unverified |
| Remoulding | EVA never enters the water | EVA goes into 70 °C water every time |
| Thickness | 4.8 mm | 3.5 mm |
| Weight | 33.3 g | 24.0 g |
| User complexity | Two parts, self-assembled | One part, ready to use |
| Unverified risk | Bond separating during play | EVA degrading after repeated heating |
Each construction has exactly one unverified weakness, and they happen to be complementary. Neither weakness showed up in this round — proving or ruling out either one needs a test built specifically around it: repeated peel testing under load for the two-piece, and repeated thermal cycling of the EVA for the one-piece.
This matters beyond material selection. It decides how your instructions are written, how many steps a player goes through, and how far a "remouldable" claim actually holds up in your product rather than in principle.
Finding 3 — 3.0 mm is below the market, not above the suppliers
We get asked for 3.0 mm often. Here is the measured picture against a real product:
| Item | Thickness | Source |
|---|---|---|
| Sample 001 (two-piece) | 4.8 mm | measured here |
| Sample 002 (one-piece) | 3.5 mm | measured here |
| Zero Ninety ZN'1 | 3.6 mm | brand's published spec |
That last row is not our measurement and we are labelling it as such — the same rule we apply to material composition applies to other people's products. Zero Ninety publish 3.6 mm for a guard carrying a carbon-reinforced shell, a foamed rubber impact layer, a lining and a grip system, at a list price around €379 in a limited run. If you are going to build a business case on a competitor's number, get it from their spec sheet on the day, not from an article — including this one.
So 3.0 mm is not a demanding target that suppliers are failing to hit. It is below the actual thickness of what is on the market. Getting there means a thinner shell plus a more expensive impact material: cost goes up, protection sits closer to the certification limit, and the wearer cannot feel the 0.5 mm.
3.5 mm is the realistic target for a version one. It puts you level with a product selling at several hundred euros, on an item that already exists and needs no tooling.
What we did not conclude
Both samples behave exactly like low-temperature thermoplastics: softening at 70 °C in 30 seconds, hand-mouldable, re-hardening on cooling, remouldable repeatedly.
We did not name the polymer.
Behaviour is consistent with a family of materials, not with one specific compound, and inferring a chemical name from how something feels in hot water is how a spec sheet ends up carrying a claim nobody can support at customs. The composition comes from the supplier's material documentation or it does not go in the report.
We would rather hand you measured behaviour than a chemical name we cannot verify.
That rule costs us a confident-sounding sentence in every report. It is worth it.
If you are speccing one of these
Three things decide the product, in this order:
- Construction — two-piece or one-piece. Decide this first; everything below it depends on the answer.
- Exit surface temperature — the safety number for a consumer-moulded product.
- Thickness target — against measured market products, not against a wish.
Send us the guard you are benchmarking against, or the spec you have been handed, and we will tell you which of the three it has already decided for you.
Want the raw data or the footage of both cycles? We filmed the whole thing — immersion, softening, moulding to the shin, cooling and the second remoulding cycle — plus the caliper and scale shots. Ask and we will send you the page.